mlfoundations-dev/Nemotron-Research-Reasoning-Qwen-1.5B_eval_5554

mlfoundations-dev/Nemotron-Research-Reasoning-Qwen-1.5Beval5554

Precomputed model outputs for evaluation.

Evaluation Results

Summary

AIME24

• —Average Accuracy: 47.67% ± 1.49%
• —Number of Runs: 10

AMC23

• —Average Accuracy: 87.50% ± 0.94%
• —Number of Runs: 10

MATH500

• —Accuracy: 86.00% | Accuracy | Questions Solved | Total Questions | |----------|-----------------|----------------| | 86.00% | 430 | 500 |

MMLUPro

• —Accuracy: 32.33% | Accuracy | Questions Solved | Total Questions | |----------|-----------------|----------------| | 32.33% | N/A | N/A |

JEEBench

• —Average Accuracy: 52.64% ± 0.55%
• —Number of Runs: 3

GPQADiamond

• —Average Accuracy: 41.75% ± 0.36%
• —Number of Runs: 3

LiveCodeBench

• —Average Accuracy: 31.38% ± 0.13%
• —Number of Runs: 3

CodeElo

• —Average Accuracy: 54.73% ± 0.39%
• —Number of Runs: 3

CodeForces

• —Average Accuracy: 40.25% ± 0.85%
• —Number of Runs: 3

HLE

• —Average Accuracy: 10.46% ± 0.27%
• —Number of Runs: 3

HMMT

• —Average Accuracy: 21.67% ± 0.85%
• —Number of Runs: 10

AIME25

• —Average Accuracy: 32.00% ± 1.35%
• —Number of Runs: 10

LiveCodeBenchv5

• —Average Accuracy: 22.58% ± 2.43%
• —Number of Runs: 3